Much controversy surrounds the inferred progenitor masses of type-II-Plateau (II-P) supernovae (SNe). The debate is nourished by the discrepant results from radiation-hydrodynamics simulations, pre-explosion imaging, and studies of host stellar populations. Here, we present a controlled experiment using four solar-metallicity models with zero-age main sequence masses of 12, 15, 20, and 25 M⊙. Because of the effects of core burning and surface mass loss, these models reach core collapse as red-supergiant (RSG) stars with a similar H-rich envelope mass of 8 to 9 M⊙ but with final masses in the range 11 to 16 M⊙. We explode the progenitors using a thermal bomb, adjusting the energy deposition to yield an asymptotic ejecta kinetic energy of 1.25 × 1051 erg and an initial 56Ni mass of 0.04 M⊙. The resulting SNe produce similar photometric and spectroscopic properties from 10 to 200 d. The spectral characteristics are degenerate. The scatter in early-time color results from the range in progenitor radii, while the differences in late-time spectra reflect the larger oxygen yields in more massive progenitors. Because the progenitors have a comparable H-rich envelope mass, the photospheric phase duration is comparable for all models; the difference in He-core mass is invisible. As different main sequence masses can produce progenitors with a similar H-rich envelope mass, light-curve modeling cannot provide a robust and unique solution for the ejecta mass of type-II-P SNe. The numerous uncertainties in massive-star evolution and wind-mass loss also prevent a robust association with a main sequence star mass. Light-curve modeling can at best propose compatibility.